Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “training”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 757 records · Page 42
All sourcesNASA NTRSDOE OSTI
DOE OSTI2019

US DOE In-Plant Trainings to Develop Expertise and Replicate Success

In-Plant Trainings (INPLTs) offered by the US Department of Energy Better Buildings, Better Plants (Better Plants) program are multi-day workshops performed by industry-recognized experts. INPLTs include both classroom- and field-based sessions that train attendees to identify energy conservation opportunities, quantify savings from these opportunities, and implement projects to realize the savings. These training workshops enhance the attendees’ understanding of working principles, knowledge of best practices, and capability to analyze the energy performance of industrial energy systems. INPLTs are competitively awarded and focus on a wide range of topics, including traditional industrial systems such as steam, process heating and motor-driven systems, and processes such as energy treasure hunts and municipal water and wastewater treatment. Partners that are awarded INPLTs can invite external stakeholders, such as employees of other manufacturers, utility representatives, and vendors. The Better Plants program had delivered 123 INPLTs as of spring 2019, trained more than 2,000 participants, and identified more than $37 million in energy cost savings. This paper provides an overview of INPLTs, some key statistics, the top identified energy savings opportunities, and success stories from INPLTs delivered from 2011 to 2018. Finally, the vision for future INPLTs is discussed based on participants’ feedback and the long-term and strategic goals of the Better Plants program.

Guo, Wei↗
DOE OSTI2022

Accelerating Collective Communication in Data Parallel Training across Deep Learning Frameworks

This work develops new techniques within Horovod, a generic communication library supporting data parallel training across deep learning frameworks. In particular, we improve the Horovod control plane by implementing a new coordination scheme that takes advantage of the characteristics of the typical data parallel training paradigm, namely the repeated execution of collectives on the gradients of a fixed set of tensors. Using a caching strategy, we execute Horovod’s existing coordinator-worker logic only once during a typical training run, replacing it with a more efficient decentralized orchestration strategy using the cached data and a global intersection of a bitvector for the remaining training duration. Next, we introduce a feature for end users to explicitly group collective operations, enabling finer grained control over the communication buffer sizes. To evaluate our proposed strategies, we conduct experiments on a world-class supercomputer — Summit. We compare our proposals to Horovod’s original design and observe 2x performance improvement at a scale of 6000 GPUs; we also compare them against tf.distribute and torch.DDP and achieve 12% better and comparable performance, respectively, using up to 1536 GPUs; we compare our solution against BytePS in typical HPC settings and achieve about 20% better performance on a scale of 768 GPUs. Finally, we test our strategies on a scientific application (STEMDL) using up to 27,600 GPUs (the entire Summit) and show that we achieve a near-linear scaling of 0.93 with a sustained performance of 1.54 exaflops (with standard error +- 0.02) in FP16 precision.

Romero, Joshua↗
DOE OSTI2021

Stable Parallel Training of Wasserstein Conditional Generative Adversarial Neural Networks : *Full/Regular Research Paper submission for the symposium CSCI-ISAI: Artificial Intelligence

We use a stable parallel approach to train Wasserstein Conditional Generative Adversarial Neural Networks (W-CGANs). The parallel training reduces the risk of mode collapse and enhances scalability by using multiple generators that are concurrently trained, each one of them focusing on a single data label. The use of the Wasserstein metric reduces the risk of cycling by stabilizing the training of each generator. We apply the approach on the CIFAR10 and the CIFAR100 datasets, two standard benchmark datasets with images of the same resolution, but different number of classes. Performance is assessed using the inception score, the Fréchet inception distance, and image quality. An improvement in inception score and Fréchet inception distance is shown in comparison to previous results obtained by performing the parallel approach on deep convolutional conditional generative adversarial neural networks (DC-CGANs). Weak scaling is attained on both datasets using up to 100 NVIDIA V100 GPUs on the OLCF supercomputer Summit.

Lupo Pasini, Massimiliano↗
DOE OSTI2024

Geometry-aware training of factorized layers in tensor Tucker format

Reducing parameter redundancies in neural network architectures is crucial for achieving feasible computational and memory requirements during train and inference of large networks. Given its easy implementation and flexibility, one promising approach is layer factorization, which reshapes weight tensors into a matrix format and parameterizes it as the product of two rank-r matrices. However, this family of approaches often requires an initial full-model warm-up phase, prior knowledge of a feasible rank, and it is sensitive to parameter initialization.In this work, we introduce a novel approach to train the factors of a Tucker decomposition of the weight tensors. Our training proposal proves to be optimal in locally approximating the original unfactorized dynamics and stable for the initialization. Furthermore, the rank of each mode is dynamically updated during training.We provide a theoretical analysis of the algorithm, showing convergence, approximation and local descent guarantees. The method's performance is further illustrated through a variety of experiments, showing remarkable training compression rates and comparable or even better performance than the full baseline and alternative layer factorization strategies.

Zangrando, Emanuele [Gran Sasso Science Institute ↗
DOE OSTI2020

Developing Subgrantee Administrative Training for the Weatherization Assistance Program

During Fiscal Years (FY) 2019 and 2020, the National Renewable Energy Laboratory (NREL) developed a series of online trainings for the U.S. Department of Energy (DOE) Weatherization Assistance Program (WAP) to support Subgrantee administrative professionals. NREL and its subcontractor developed a series of online administrative trainings for WAP Subgrantees. In total, 29 administrative training courses were developed for WAP Subgrantees, focused on the following job roles: Weatherization Director, Fiscal Manage, Quality Assurance Manager. As of September 8, 2020, 2,840 users in nearly all WAP Grantee jurisdictions had registered with the learning management system and completed over 23,000 courses. This report provides a summary of the development process, rollout, and outreach to the WAP network, user response, and lessons learned.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Multilevel-in-width training for deep neural network regression

A common challenge in regression is that for many problems, the degrees of freedom required for a high-quality solution also allows for overfitting. Regularization is a class of strategies that seek to restrict the range of possible solutions so as to discourage overfitting while still enabling good solutions, and different regularization strategies impose different types of restrictions. In this paper, we present a multilevel regularization strategy that constructs and trains a hierarchy of neural networks, each of which has layers that are wider versions of the previous network's layers. Here, we draw intuition and techniques from the field of Algebraic Multigrid (AMG), traditionally used for solving linear and nonlinear systems of equations, and specifically adapt the Full Approximation Scheme (FAS) for nonlinear systems of equations to the problem of deep learning. Training through V-cycles then encourage the neural networks to build a hierarchical understanding of the problem. We refer to this approach as multilevel-in-width to distinguish from prior multilevel works which hierarchically alter the depth of neural networks. The resulting approach is a highly flexible framework that can be applied to a variety of layer types, which we demonstrate with both fully connected and convolutional layers. We experimentally show with PDE regression problems that our multilevel training approach is an effective regularizer, improving the generalize performance of the neural networks studied.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2013

The Importance of Curriculum-Based Training and Assessment in Interventional Radiology

Physician performance and outcomes are being scrutinised by health care providers to improve patient safety and cost efficiency. Patients are best served by physicians who have undergone appropriate specialist training and assessment and perform large numbers of cases to maintain their skills. The Cardiovascular and Interventional Radiological Society of Europe has put into place a curriculum for training in interventional radiology (IR) and a syllabus with an examination, the European Board of Interventional Radiology, providing evidence of attainment of an appropriate and satisfactory skill set for the safe practice of IR. This curriculum is appropriate for IR where there is a high volume of image-guided procedures in vascular and nonvascular organ systems with cross-use of minimally invasive techniques in patients with a variety of disease processes. Other specialties may require different, longer, and more focused training if their experience is “diluted” by the need to master a different skill set.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2015

Identifying the Learning Curve for Uterine Artery Embolisation in an Interventional Radiological Training Unit

AimA clear understanding of operator experience is important in improving technical success whilst minimising patient risk undergoing endovascular procedures, and there is the need to ensure that trainees have the appropriate skills as primary operators. The aim of the study is to retrospectively analyse uterine artery embolisation (UAE) procedures performed by interventional radiology (IR) trainees at an IR training unit analysing fluoroscopy times and radiation dose as surrogate markers of technical skill.MethodsTen IR fellows were primary operator in 200 UAE procedures over a 5-year period. We compared fluoroscopy times, radiation dose and complications, after having them categorised according to three groups: Group 1, initial five, Group 2, >5 procedures and Group 3, penultimate five UAE procedures. We documented factors that may affect screening time (number of vials employed and use of microcatheters).ResultsMean fluoroscopy time was 18.4 (±8.1), 17.3 (±9.0), 16.3 (±8.4) min in Groups 1, 2 and 3, respectively. There was no statistically significant difference between these groups (p > 0.05) with respect to fluoroscopy time or radiation dose. Analysis after correction for confounding factors showed no statistical significance (p > 0.05). All procedures were technically successful, and total complication rate was 4 %.ConclusionUAE was chosen as a highly standardised procedure followed by IR practitioners. Although there is a non-significant trend for shorter screening times with experience, technical success and safety were not compromised with appropriate Consultant supervision, which illustrates a safe construct for IR training. This is important and reassuring information for patients undergoing a procedure in a training unit.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2018

A continuous analogue of the tensor-train decomposition

We develop new approximation algorithms and data structures for representing and computing with multivariate functions using the functional tensor-train (FT), a continuous extension of the tensor-train (TT) decomposition. The FT represents functions using a tensor-train ansatz by replacing the three-dimensional TT cores with univariate matrix-valued functions. The main contribution of this paper is a framework to compute the FT that employs adaptive approximations of univariate fibers, and that is not connected to any tensorized discretization. The algorithm can be coupled with any univariate linear or nonlinear approximation procedure. We demonstrate that this approach can generate multivariate function approximations that are several orders of magnitude more accurate, for the same cost, than those based on the conventional approach of compressing the coefficient tensor of a tensor-product basis. Our approach is in the spirit of other continuous computation packages such as Chebfun, and yields an algorithm which requires the computation of “continuous” matrix factorizations such as the LU and QR decompositions of vector-valued functions. To support these developments, we describe continuous versions of an approximate maximum-volume cross approximation algorithm and of a rounding algorithm that re-approximates an FT by one of lower ranks. We demonstrate that our technique improves accuracy and robustness, compared to TT and quantics-TT approaches with fixed parameterizations, of high-dimensional integration, differentiation, and approximation of functions with local features such as discontinuities and other nonlinearities.

42 ENGINEERING↗
DOE OSTI2014

Establishing High-Quality Prostate Brachytherapy Using a Phantom Simulator Training Program

Purpose: To design and implement a unique training program that uses a phantom-based simulator to teach the process of prostate brachytherapy (PB) quality assurance and improve the quality of education. Methods and Materials: Trainees in our simulator program were practicing radiation oncologists, radiation oncology residents, and fellows of the American Brachytherapy Society. The program emphasized 6 core areas of quality assurance: patient selection, simulation, treatment planning, implant technique, treatment evaluation, and outcome assessment. Using the Iodine 125 ({sup 125}I) preoperative treatment planning technique, trainees implanted their ultrasound phantoms with dummy seeds (ie, seeds with no activity). Pre- and postimplant dosimetric parameters were compared and correlated using regression analysis. Results: Thirty-one trainees successfully completed the simulator program during the period under study. The mean phantom prostate size, number of seeds used, and total activity were generally consistent between trainees. All trainees met the V100 >95% objective both before and after implantation. Regardless of the initial volume of the prostate phantom, trainees' ability to cover the target volume with at least 100% of the dose (V100) was not compromised (R=0.99 pre- and postimplant). However, the V150 had lower concordance (R=0.37) and may better reflect heterogeneity control of the implant process. Conclusions: Analysis of implants from this phantom-based simulator shows a high degree of consistency between trainees and uniformly high-quality implants with respect to parameters used in clinical practice. This training program provides a valuable educational opportunity that improves the quality of PB training and likely accelerates the learning curve inherent in PB. Prostate phantom implantation can be a valuable first step in the acquisition of the required skills to safely perform PB.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2025

The tensor-train stochastic finite volume method for uncertainty quantification

The stochastic finite volume method offers an efficient one-pass approach for assessing uncertainty in hyperbolic conservation laws. Still, it struggles with the curse of dimensionality when dealing with multiple stochastic variables. Here, we introduce the stochastic finite volume method within the tensor-train framework to counteract this limitation. This integration, however, comes with its own set of difficulties, mainly due to the propensity for shock formation in hyperbolic systems. To overcome these issues, we have developed a tensor-train-adapted stochastic finite volume method that employs a global WENO reconstruction, making it suitable for such complex systems. This approach represents the first step in designing tensor-train techniques for hyperbolic systems and conservation laws involving shocks.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Fast neural network training on a cluster of GPUs for action recognition with high accuracy

In this work, we propose algorithms and techniques to accelerate training of deep neural networks for action recognition on a cluster of GPUs. The convergence analysis of our algorithm shows it is possible to reduce communication cost and at the same time minimize the number of iterations needed for convergence. We customize the Adam optimizer for our distributed algorithm to improve efficiency. In addition, we employ transfer-learning to further reduce training time while improving validation accuracy. For the UCF101 and HMDB51 datasets, the validation accuracies achieved are 93.1% and 67.9% respectively. With an additional end-to-end trained temporal stream, the validation accuracies achieved for UCF101 and HMDB51 are 93.47% and 81.24% respectively. As far as we know, these are the highest accuracies achieved with the two-stream approach using ResNet that does not involve computationally expensive 3D convolutions or pretraining on much larger datasets.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Longitudinal train dynamics model for a rail transit simulation system

The paper develops a longitudinal train dynamics model in support of microscopic railway transportation simulation. The model can be calibrated without any mechanical data making it ideal for implementation in transportation simulators. The calibration and validation work is based on data collected from the Portland light rail train fleet. The calibration procedure is mathematically formulated as a constrained non-linear optimization problem. The validity of the model is assessed by comparing instantaneous model predictions against field observations, and also evaluated in the domains of acceleration/deceleration versus speed and acceleration/deceleration versus distance. A test is conducted to investigate the adequacy of the model in simulation implementation. The results demonstrate that the proposed model can adequately capture instantaneous train dynamics, and provides good performance in the simulation test. Thus, the model provides a simple theoretical foundation for microscopic simulators and will significantly support the planning, management and control of railway transportation systems.

42 ENGINEERING↗
DOE OSTI2024

MedNAS: Multiscale Training-Free Neural Architecture Search for Medical Image Analysis

Deep neural networks have demonstrated impressive results in medical image analysis, but designing suitable architectures for each specific task is expertise dependent and time consuming. Neural architecture search (NAS) offers an effective means of discovering architectures. It has been highly successful in numerous applications, particularly in natural image classification. Yet, medical images possess unique characteristics, such as small regions and a wide variety of lesion sizes, that differentiate them from natural images. Furthermore, most current NAS methods struggle with high computational costs, especially when dealing with high-resolution image datasets. In this article, we present a novel evolutionary NAS method called multiscale training-free neural architecture search (MSTF-NAS) to address these challenges. Specifically, to accommodate the broad range of lesion region sizes in disease diagnosis, we develop a new reduction cell search space that enables the search algorithm to explicitly identify the optimal scale combination for multiscale feature extraction. Further, to overcome the issue of high computational costs, we utilize training-free indicators as performance measures for candidate architectures, which allows us to search for the optimal architecture more efficiently. More specifically, by considering the capability and simplicity of various networks, we formulate a multiobjective optimization problem that involves two training-free indicators and model complexity for candidate architectures. Extensive experiments on a large medical image benchmark and a publicly available breast cancer detection dataset are conducted. The empirical results demonstrate that our MSTF-NAS outperforms both human-designed architectures and current state-of-the-art NAS algorithms on both datasets, indicating the effectiveness of our proposed method.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

SU-E-E-05: Improving Contouring Precision and Consistency for Physicians-In-Training with Simple Lab Experiments

Purpose: Target contouring for high-dose treatments such as radiosurgery of brain metastases is highly critical in eliminating marginal failure and reducing complications as shown by recent clinical studies. In order to improve contouring accuracy and practice consistency for the procedure, we introduced a self-assessed physics lab practice for the physicians-in-training. Methods: A set of commercially acquired high-precision PMMA plastic spheres were randomly embedded in a Styrofoam block and then scanned with the CT/MR via the clinical procedural imaging protocol. A group of first-year physicians-in-training (n=6) from either neurosurgery or radiation oncology department were asked to contour the scanned objects (diameter ranged from 0.4 cm to 3.8 cm). These user-defined contours were then compared with the ideal contour sets of object shape for self assessments to determine the maximum areas of the observed discrepancies and method of improvements. Results: The largest discrepancies from initial practice were consistently found to be located near the extreme longitudinal portions of the target for all the residents. Discrepancy was especially prominent when contouring small objects < 1.0 cm in diameters. For example, the mean volumes rendered from the initial contour data set differed from the ideal data set by 7.7%±6.6% for the participants (p> 0.23 suggesting agreement cannot be established). However, when incorporating a secondary imaging scan such as reconstructed coronal or sagittal images in a repeat practice, the agreement was dramatically improved yielding p<0.02 in agreement with the reference data set for all the participants. Conclusion: A simple physics lab revealed a common pitfall in contouring small metastatic brain tumors for radiosurgical procedures and provided a systematic tool for physicians-in-training in improving their clinical contouring skills. Dr Ma is current a board member of international stereotactic radiosurgical society.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

MO-DE-BRA-04: Hands-On Fluoroscopy Safety Training with Real-Time Patient and Staff Dosimetry

Purpose: Fluoroscopically guided interventions (FGI) are routinely performed across many different hospital departments. However, many involved staff members have minimal training regarding safe and optimal use of fluoroscopy systems. We developed and taught a hands-on fluoroscopy safety class incorporating real-time patient and staff dosimetry in order to promote safer and more optimal use of fluoroscopy during FGI. Methods: The hands-on fluoroscopy safety class is taught in an FGI suite, unique to each department. A patient equivalent phantom is set on the patient table with an ion chamber positioned at the x-ray beam entrance to the phantom. This provides a surrogate measure of patient entrance dose. Multiple solid state dosimeters (RaySafe i2 dosimetry systemTM) are deployed at different distances from the phantom (0.1, 1, 3 meters), which provide surrogate measures of staff dose. Instructors direct participating clinical staff to operate the fluoroscopy system as they view live fluoroscopic images, patient entrance dose, and staff doses in real-time. During class, instructors work with clinical staff to investigate how patient entrance dose, staff doses, and image quality are affected by different parameters, including pulse rate, magnification, collimation, beam angulation, imaging mode, system geometry, distance, and shielding. Results: Real-time dose visualization enables clinical staff to directly see and learn how to optimize their use of their own fluoroscopy system to minimize patient and staff dose, yet maintain sufficient image quality for FGI. As a direct result of the class, multiple hospital departments have implemented changes to their imaging protocols, including reduction of the default fluoroscopy pulse rate and increased use of collimation and lower dose fluoroscopy modes. Conclusion: Hands-on fluoroscopy safety training substantially benefits from real-time patient and staff dosimetry incorporated into the class. Real-time dose display helps clinical staff visualize, internalize, and ultimately utilize the safety techniques learned during the training. RaySafe/Unfors/Fluke lent us a portable version of their RaySafe i2 Dosimetry System for 6 months.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

MO-DE-BRA-05: EUTEMPE-RX: Combining E-Learning and Face-To-Face Training to Build Expert Knowledge, Skills and Competences for Medical Physicists in Diagnostic and Interventional Radiology

Purpose: In 2013, the EURATOM authorities of the European Commission decided to support the Horizon2020 project submission ‘EUTEMPE-RX’ that aimed for a new set of course modules to train medical physicists in diagnostic and interventional radiology to expert level with small group deep learning. Each module would consist of 2 phases: an e-learning and a face-to-face phase, each phase requiring typically 40h of participant time. Methods: The European Federation (EFOMP) and 13 European partners, all of them selected for their excellent scientific and/or educational skills, led the 12 course modules. A quality manual ensured the quality of course content and organization. Educational workshops familiarized the teachers with e-learning techniques and methods for assessment. Content was set in accordance with the EC document RP174 that lists learning outcomes in terms of knowledge, skills and competences (KSCs) for different specialties and levels of medical physics. Surveys for stake holder satisfaction were prepared. Results: Today the course modules are being realized. The modules cover most of the KSCs in RP174 document. Teachers have challenged the participants with unique tasks: case studies in medical physics leadership, Monte Carlo simulation of a complete x-ray imaging chain, development of a task specific QA protocol, compilation of optimization plans, simulation tasks with anthropomorphic breast models, etc. Participants undertook practical sessions in modern hospitals and visited a synchrotron facility, a calibration lab, screening organizations, etc. Feedback form quality surveys was very positive and constructive. A sustainability plan has been worked out. Conclusion: The modules have enabled the participants to develop their KSCs and cope with challenges in medical physics. The sustainability plan will be implemented to continue the unique combined e-learning and face to face training at high level training in diagnostic and interventional radiology also beyond Europe. The EUTEMPE-RX project is funded by the Horizon2020 - EURATOM program of the European Commission.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

WE-G-204-00: Post-Graduate Training of the Next Generation of Academic Medical Physicists

While many indicators for academic medical physics are distressing – jobs are tight, demands on clinical time are high (and getting worse) and national funding has been flat for several years (meaning less money in reality) the present is perhaps one of the most exciting times in cancer research history, and medical physicists have an opportunity to make a difference. Many of us predict the impact of medical physics on cancer research over the next decade to be more significant than ever. Why is that? First, medical imaging is used for every cancer patient in developed countries. Every improvement in the acquisition, processing or analysis of radiological images has the potential to impact patients. The use of radiation therapy is at an all-time high – and virtually cannot be performed without medical physics. Many of the advances in both biomedical imaging and radiation oncology are the result of the hard work of academic medical physicists who are thinking of the next generation of technologies that will be used against cancer or an even broader spectrum of diseases. A career in academic medical physics is demanding, particularly for those with clinical responsibilities. As the demands for justification of their clinical effort become increasingly metricized, the ability to do “unfunded research” will become even more difficult. This means that many will have to generate external salary support to justify their efforts in research and development. This comes at a time when funding for research is compressed and harder to obtain. Generally speaking, if you are not contributing 50% or more of your effort to research, you are competing at a disadvantage and it is very unlikely you will get an NIH/NCI/NIBIB grant. Furthermore, in the ongoing effort to improve patient care and safety, we have developed credentialing pathways that now require at least two-years of residency training. This full-time clinical training creates a gap in the research trajectory of graduate students who aspire to academic positions with an expectation for extramural funding. To address this, several residency programs have created hybrid programs where the two-years of clinical training is combined with one or two years of research effort to allow candidates to further establish an academic identity and to ensure adequate academic productivity to compete for a beginning faculty position. In conclusion, while the path to a successful career in academic medical physics is steep and sometimes hard to follow, reaching the apex is worth the journey. Different paths to a career in medical physics are available, you just have to decide which one is right for you. If improving cancer care is your goal as a physicist, then academic medical physics is the job for you!.

60 APPLIED LIFE SCIENCES↗